The group’s scientific interest is particularly oriented on the redox chemistry of transition metal (TM) complexes, especially their multi-electron reactivity, photoredox properties and their application in small molecule activation. Our profound goal is to understand how to rationally design two- and multi-electron reactivity of redox cooperative catalysts to efficiently and catalytically transform abundant but inert feedstock molecules (mostly N2, CO2, H2 and H2O) to valuable chemicals. In addition, we wish to gain explicit control over the redox energetics of robust photoredox catalysts and rationally design their redox properties, especially towards highly energy demanding net reductions.
One critical mission of the group is to show and convince the broad research community that the design redox cooperative and photoredox catalysts and redox leveling of TM complexes in general can be revolutionized. With our studies we want to demonstrate that careful analysis of computed results can lead to intuitively understandable concepts that can be generalized and applied, sometimes in a quantitative manner, to predict important properties including redox potentials and multi-electron reactivity. The large-scale computational approach that we have been experimenting and perfecting in the last several years is opening new horizons in utilizing first principle quantum chemical simulations to effectively guide and support experimental studies.
The ongoing research topics of the group are (but not limited to):
- redox-active and non-innocent ligands
- spin-state energetics and spin-crossover of TM complexes
- reactivity of M=X and M≡X (X = C-R, N) functionalities
- secondary chemical interactions
- data management, DOMINO
REDOX-ACTIVE AND NON-INNOCENT LIGANDS
Our focus and expertise have been centered at redox-active and non-innocent ligands and their application as electron reservoirs in transition metal complexes. Within this project introducing we introduced, amongst others, a novel concept that accounts for the high electron accepting ability of the common X=C–C=Y structural motif (X, Y = O, and NH) of ligands, shared for example by quinone and bipridine ligands and other archetype redox-active scaffolds. By means of a systematic redox-potential investigation on an extensive set of transition metal complexes we also revealed the most critical factors, such as ring fusion, nature of contact atoms, substitution, etc. that influence the redox energetics of ligand-centered electron transfer events in TM complexes. We have established an innovative DFT-based method for scrutinizing the effective oxidation state of the metals and ligands in transition metal complexes containing non-innocent ligands. We are currently working on applying our ligand design strategies and large-scale computational approach to establish innovative concepts in designing the multi-electron reactivity of redox cooperative catalysts that utilize redox-active ligands as electron reservoirs. Within the scope of this research line we also studied the potential effect of redox activity of extended porphyrins on their Möbius and Hückel aromaticity in d8 TM complexes. Moreover, in our ongoing collaborative research we are scrutinizing the critical importance of redox active behavior of porphyrins in determining the activity of P450 enzymes towards breaking inert C–H bonds through a QM/MM study.
SPIN-STATE ENERGETICS AND SPIN-CROSSOVER
Computing and predicting consistent spin state energetics in general is a challenging task, especially when using Density Functional Theory, which is known to give chronically inconsistent spin-state energetics. As a matter of fact, the inability of computing consistent spin-state energetics remains a persistent and global problem of computational chemistry that seriously limits theory to support or perhaps guide experimental research for many 1st-row transition metal complexes. Within this collaborative research we aim at providing conceptual insight and physical reasons for the inherently inconsistent DFT spin-state energetics and its dependence on the exact exchange admixture of the functional. Also, as our perception is that the recently developed particle-particle Random Phase Approximation (pp-RPA) might potentially become a unique remedy for the failure of DFT to describe electron correlation in TM complexes, in collaboration with Prof. Yang we are currently working on developing a novel pp-RPA-DFT approach for computing the spin-state energetics and MLCT excitations of transition metal complexes. In collaboration with Prof. Neese (MPI-CEC), we also started revealing the effect of metal-based spin-crossover (SCO) processes on the redox activity of ligands in iron and cobalt complexes using a benchmarked computational protocol that provides satisfactory accuracy for these iron complexes exhibiting SCO and redox active ligands in the same time. These projects are directly related to the scope of the European COST action (ECOSTBio) in which we are participating.
REACTIVITY OF M=X and M≡X (X = C-R, N) FUNCTIONALITIES
We very much like to participate in interdisciplinary research and contribute with our expertise in theoretical/conceptual/computational chemistry to experimental-driven research. We have a long-standing fruitful collaboration with Prof. Mindiola (UPenn, USA), Prof. Košmrlj (University of Ljubljana, Slovenia) and Prof. Fortier (University of Texas at El Paso, USA) focusing on the redox processes, small molecules activation, and other aspects of organometallic and inorganic species. Through these combined theoretical/experimental studies, we provided a detailed scrutiny for and conceptualized the reactivity of Ti≡C–R functionalities (R = tBu, PPh2) towards methane, ethane, other linear and cyclic alkanes, benzene and ethers, while more recently we reported on the nucleophilic reactivity of titanium-nitrides and on the structure of novel terminal zirconium and hafnium methylidenes. We also investigated the thermodynamic and kinetic aspects of N2 activation by d3 TM species and revealed the most critical structural and electronic requirements needed to break the N≡N bond with such complexes. Our investigation on cyclo-P3 complexes also showcased how to intuitively conceptualize the chemical shielding of heavy nuclei within Ramsey’s theory in order to gain first-hand information about the nature and characteristics of metal-ligand interactions. We also gained exclusive skills through studies aiming at the computational and experimental characterization of the ‘click chemistry’ of inorganic complexes, dinitrogen liberation and the mechanism of copper-free Sonogashira reaction.
SECONDARY CHEMICAL INTERACTIONS
The theoretical investigation of secondary chemical interactions is also a recurring feature of the research activity of the group. We have a great deal of expertise in several state-of-the-art analyzing techniques, such as energy decomposition analyses (e.g. EDA), Natural Orbital for Chemical Valence (NOCV) analysis, Non-Covalent Interaction (NCI) method, Natural Resonance Theory (NRT), MEP, aromaticity measures, Fukui functions and QTAIM. Applying a balanced combination of these techniques provided novel insights already into the nature and strength of different d10 M…M-type metallophilic interactions (M = Cu, Ag, Au), halogen-bonding and its competition with hydrogen-bonding in elegantly designed molecules, lone-pair…π* interactions, chalcogen-bonding, ion-π and ion-σ interactions, as well as it helped to reveal the effect of various weak interactions, such as π-π stacking, C–H…π and C–H…F interactions, to the formation of frustrated Lewis pairs (FLPs) and to their characteristic reactivity towards molecular hydrogen.
DOMINO, the Big Data management solution
Our main research is heavily driven by first principle computational simulations carried out in a large-scale fashion. We most often explore the chemical space at an unprecedentedly high resolution, test methods, benchmark data and we also put a great emphasis on accumulating these consistent raw data into an organized database.
We created DOMINO, a well-designed data managing system with an easy-to-follow user-friendly interface to keep track of the performed simulations. The generated raw data is systematically collected, saved securely in a database that can be easily queried. DOMINO not only does all these above, but it is made to reduce the possibility of making the most common errors in setting up calculations in a controllable way. It facilitates the workflow from input generation through file management and transfer to HPC, data/results analyzing, up till data storing and quick and easy data sharing between peers, supervisors, and collaborators. It simplifies and speeds up the unavoidable manual work that is associated with quantum chemical calculations. With DOMINO we wish to facilitate and promote an effective work and set best practices for doing quantum chemical calculations for large-scale projects that wish to go beyond standard quantum chemistry studies.
We realize that all the data that we generate is and will be highly valuable in the future for other projects therefore DOMINO collects these data is a systematic, clear, searchable database. In time we wish to link these datasets from the various projects to gain new insights and perhaps new ideas by the aid of machine learning. It is important to realize, that not the quantity of data that is revolutionary in our concept, but its quality and the things that we can do with it. Big Data’s potential benefits to research and society go far beyond what has been accomplished so far.
While we build our science on data coming out from computers, we must emphasize that you don’t get good scientific output from throwing everything against the wall and seeing what sticks. No matter how much data one generates, we still need to ask the right questions to create a hypothesis, design a test, and use the data to determine whether that hypothesis is true!